Nvidia faces gamer backlash as community feels abandoned in the AI gold rush
DLSS 5, generative AI, increasingly expensive GPUs, and a gaming community that feels left behind. That is the reality Nvidia is dealing with right now, after three decades of being practically synonymous with gaming.
The company that nearly went bankrupt in 1999, saved by the very gamers who bought its first graphics cards, now dominates headlines about billions in AI contracts, data center GPUs costing up to 40 thousand dollars a unit, and a consumer hardware roadmap that, for the first time in 30 years, might not include a single new generation in 2026.
And then came the phrase that summed it all up better than any earnings report ever could.
Dance with the one who brought you. That was Greg Miller, co-founder of the Kinda Funny Games Daily podcast, speaking to CNBC, followed by a simple and blunt confession: it breaks my heart.
But is Nvidia really abandoning gamers, or has the market simply shifted too fast for anyone to keep up? There is a lot going on beneath the surface 👇
AI profits took over everything
Nvidia’s era of AI dominance did not start yesterday. It all goes back to the launch of CUDA, its software toolkit, in 2006. From that point on, developers could use GPUs for general-purpose computing, not just graphics. The moment many consider the big bang of modern AI came in 2012, when Nvidia GPUs and CUDA were used to build a neural network called AlexNet, which crushed the competition in an image recognition contest.
In 2020, Nvidia signaled a new strategic focus by acquiring Mellanox Technologies, a maker of chips for high-performance computing, for 7 billion dollars. Since then, the company has been releasing generations of ultra-high-performance GPUs built for AI, along with complete rack-scale systems like the Vera Rubin platform, which CNBC got exclusive access to in February.
The numbers explain the shift in priorities pretty clearly. Operating margins for Nvidia’s compute and networking segment averaged 69% over the last three years. The graphics segment, which serves consumers, came in at 40%. A Blackwell GPU for data centers can cost up to 40 thousand dollars, and the Futurum Group estimates a full Vera Rubin system could run as high as 4 million dollars. For comparison, the RTX 50 series for gamers is priced between 299 and 1,999 dollars.
Stacy Rasgon of Bernstein Research put it plainly: the gaming segment is no longer the driving force behind the company. There was a point in time when it clearly was, but that time has passed.
What DLSS 5 has to do with all of this
CEO Jensen Huang made a gaming-focused announcement at the top of his keynote at GTC 2026 in March, revealing the next generation of the Deep Learning Super Sampling rendering software, DLSS 5, expected to launch in the second half of the year. The technology is known for boosting frame rates by rendering games at lower resolutions and using AI to upscale the image, helping titles run more smoothly on less powerful hardware.
The controversy came from the fact that DLSS 5 uses generative AI to create entirely new frames that did not exist in the original rendering pipeline. Huang introduced the feature with a video showing photorealistically enhanced versions of characters from popular games like Resident Evil Requiem, Starfield, and Hogwarts Legacy. And that is when the gaming community pushed back hard.
Miller from Kinda Funny Games was blunt: I play video games because it is an art form. I like to see the fingerprint of the creator in what I am playing. He said the demo raised a lot of red flags across the games industry, especially at a time when the sector is dealing with so many layoffs and studio closures.
Tim Gettys, co-founder of Kinda Funny Games alongside Miller, was a fan of previous DLSS versions for making games more accessible on a budget. But the addition of the generative layer changed his take: the technology is impressive for what it does to run games on more modest PCs. But putting this generative AI stuff on top of it feels like a slap in the face.
Receive the best innovation content in your email.
All the news, tips, trends, and resources you're looking for, delivered to your inbox.
By subscribing to the newsletter, you agree to receive communications from Método Viral. We are committed to always protecting and respecting your privacy.
Gettys’s bigger fear is that this is a stepping stone toward fully AI-generated games, something he believes is the end goal. Elon Musk has already raised that possibility publicly. In October, Musk posted on X that his xAI game studio would release a major AI-generated game before the end of 2026.
Gettys was direct: you are literally altering the art created by developers. And at some point you are replacing the developers, and then their studio shuts down.
Huang responded to the criticism during a Q&A session the day after the GTC keynote. He said the claims that DLSS 5 makes games look homogeneous are completely wrong and emphasized that developers will remain in control, able to tune the generative AI to match their artistic style.
Nvidia also told CNBC that games are a form of creative art and that its RTX technologies, including ray tracing, path tracing, DLSS Super Resolution, DLSS Frame Generation, and DLSS 5, are tools for developers to achieve their creative vision while delivering the best possible performance and image quality.
The memory crisis behind it all
There is a technical and economic factor that most gamers are not fully aware of, and it explains a big chunk of the problem. A DRAM shortage is at the center of the issue.
Industry reports indicate that Nvidia plans to cut production of its latest gaming GPUs by up to 40% because of a major shortage of the general-purpose memory needed to manufacture a GPU. DRAM (Dynamic Random Access Memory) enables fast, temporary data storage so the GPU can handle parallel tasks.
Personal PCs, where Nvidia’s gaming GPUs are installed, have taken the hardest hit from this shortage. When memory prices go up, building a GPU costs more, and that cost gets passed down to the end consumer. Gartner predicts PC prices will climb 17% this year, causing global shipments to drop 10.4%.
Gettys weighed in on the situation: with everything getting so expensive, it is concerning to see prices going up on the gaming side with no sign of coming back down, while Nvidia is clearly chasing an entirely different category of customer.
The technical explanation is revealing. High-performance GPUs like Blackwell and Rubin use dense stacks of a specific type of DRAM called HBM (High Bandwidth Memory). According to Rasgon at Bernstein, it takes roughly four times as many silicon wafers to produce one gigabyte of HBM compared to the same amount of traditional DRAM.
That dynamic is starving the entire industry of the kind of memory traditionally used for more consumer-facing applications. It is just not available, Rasgon said. If there are delays on the gaming roadmap, it is largely because they probably cannot even build the cards, since the memory is hard to come by. Every bit of memory that exists is being prioritized for AI compute.
Nvidia told CNBC that it continues shipping all GeForce GPUs, that it sees strong demand, and that it is working closely with suppliers to maximize memory availability.
The Nvidia that gamers built, and the one they lost
There is some historical context here that really changes how you look at the current moment. In 1999, Nvidia was on the verge of financial collapse. The company had bet everything on a GPU called NV10, which would become the GeForce 256, and its survival depended on that launch. The company laid off most of its employees and came close to bankruptcy to make the release happen. It was the gaming community that bought those cards and kept the company alive. Not data centers. Not AI researchers. Gamers.
That relationship deepened over the following decades. Nvidia popularized graphics processing units, GPUs, which enable fast frame rates and high-quality rendering for the best possible gaming experience.
If analyst predictions hold up, 2026 will be the first year in three decades without the launch of a new generation of the consumer-facing GeForce lineup. The RTX 50 series was unveiled at CES in January 2025, and with CES 2026 and GTC already in the rearview mirror, many fear this will be the first year without a generational release, although Nvidia has historically revealed new hardware as late as September.
The company told CNBC via email that gamers are extremely important to them and that they are always innovating, testing, and releasing new technologies focused on gaming.
Not everyone sees the gap as a bad thing. Gettys noted: it is kind of hard to keep up. You cannot upgrade every year, so having a breather and waiting for a generation that really makes a difference I think actually benefits gamers.
Sky-high prices and the crypto mining hangover
The pricing issue is nothing new for Nvidia. During the cryptocurrency booms of 2018 and 2021, the company’s GPUs were selling on online marketplaces for up to three times their retail price because they were essential for mining Bitcoin and Ethereum. While prices came down after mining shifted direction in 2022, the current RTX 5090 is still being sold online for up to double the suggested retail price.
That elevated demand for the current generation could make Nvidia less motivated to launch a new version this year. After all, if inventory sells out even at inflated prices, what is the immediate financial incentive to speed up the next cycle?
Data from Steam hardware surveys shows that the most popular GPU among the platform’s users is still a budget card released years ago. That says a lot about where most gamers are in the real world, far away from RTX 5090s and conversations about DLSS 5 with multi-frame generation. The discussion around AI in games is happening within a slice of the community that represents a small fraction of total active players.
If the entry-level PC market disappears by 2028, as Gartner predicts, the market for Nvidia’s budget gaming GPUs is likely to shrink along with it.
The games industry in crisis and the weight of layoffs
The broader games industry context makes everything even more sensitive. While dealing with a post-pandemic slowdown, the sector has seen a wave of studio closures, canceled games, and thousands of layoffs at giants like Epic Games, Microsoft’s Xbox, and Sony’s PlayStation.
When Nvidia presents a technology that, in the eyes of gamers, uses generative AI to alter art created by human developers, the timing could not be worse. Miller captured that sentiment when he said the DLSS 5 demo raised a lot of hackles in the video game industry, right at a moment when the sector is dealing with so many layoffs and studio shutdowns.
The fear that generative AI will eventually replace human developers is real and deeply felt within the community. Gettys was clear: if they are making three times as much money and shareholders are three times as happy, then yeah, I think they will abandon gaming even though that is what brought them here.
GeForce NOW and the bright side of the story
Not everything is bad news in the relationship between Nvidia and gamers. For over a decade, the company has offered cloud gaming through its GeForce NOW service. The model has evolved to include different subscription tiers, including a free option, which lets users stream games they already own on platforms like Steam, running on Nvidia GPUs in data centers instead of on personal devices.
Miller praised the platform: you see Xbox and PlayStation and other competitors trying to bring the cloud to gamers in a way that makes sense. And Nvidia with GeForce NOW really cracked the code.
Gettys was even more emphatic, calling Nvidia’s streaming platform the best by far. It allows millions more people to access gaming at the highest level, even without the latest cards. It is truly incredible technology.
This is perhaps the clearest path Nvidia has to keep gamers engaged even without releasing new hardware every year. If cloud infrastructure lets anyone play at top-tier quality without needing an RTX 5090 on their desk, GeForce NOW could become the bridge between the company’s gaming era and its AI era.
AMD and the competition that is also struggling
AMD, with its Radeon line of GPUs, is Nvidia’s main competitor in the gaming segment. But the memory crisis is hitting both companies.
Rasgon left no room for doubt: if Nvidia cannot get the memory, AMD is not going to get it either. He added that in terms of brand sentiment, both have dedicated fans who can be pretty passionate.
Still, Gettys pointed to a clear market preference: there is an obvious favorite. If you game on PC, you are going to want an Nvidia card.
That market dynamic means that even with all the frustration, PC gamers do not have many real alternatives. Intel is trying to compete with its Arc GPUs, but it has not come close to threatening Nvidia or AMD’s position in the high-performance segment. That gives Nvidia a comfortable margin to prioritize its AI chips without losing significant gaming market share, at least in the short term.
What gamers actually want from Nvidia
The answer is less complicated than it seems. The gaming community is not asking Nvidia to abandon AI or stop investing in cutting-edge technology. What most players want is a clear signal that there is still a place for them in the company’s strategy, not just as second-class consumers getting watered-down versions of technologies designed for data centers, but as the audience that historically drove Nvidia’s innovation in graphics processing.
That translates into some pretty concrete things: GPUs priced in a range that makes sense for the average gamer’s budget, transparent communication about the future hardware roadmap, and access to technologies like DLSS 5 on entry-level and mid-range cards within a reasonable timeframe.
Greg Miller’s phrase, dance with the one who brought you, resonated so widely because it captures that feeling exactly. It is not anger. It is disappointment from people who backed a company when it was small and vulnerable, and now watch that same company look the other way while counting its billions.
Historical trends show that cutting-edge technologies eventually trickle down to more affordable tiers of the market. Ray tracing, which in 2018 only ran acceptably on the RTX 2080 and 2080 Ti, now works well on budget GPUs like the RTX 4060. DLSS 5 and AI-powered frame generation will likely follow the same path over the next hardware cycles. But until that happens, a huge portion of gamers is left on the sidelines, and the feeling of abandonment has everything to do with that gap between when the technology launches and when it actually becomes accessible to everyone.
It could be that the market has shifted in a way that cannot be reversed and that Nvidia is simply adapting to the new reality. But the way that adaptation is being communicated, or rather the lack of direct communication about it, is what keeps feeding the feeling that gamers have been pushed to the back of the line. 🎮
